Title :
Corrective training of hidden control neural network
Author :
Na, KyungMin ; Chae, Soo-Ik ; Ann, Souguil
Author_Institution :
Dept. of Electron. Eng., Seoul Nat. Univ., South Korea
Abstract :
A corrective training algorithm for hidden control neural network (HCNN) is proposed in this paper with application to the isolated spoken Korean digit recognition. The proposed algorithm tries to heuristically minimize the number of recognition errors, which improves the discriminatory power of the conventional HCNN-based speech recognizers. Experimental results showed 25% reduction for closed test, and 10% reduction for open test in the number of recognition errors
Keywords :
heuristic programming; minimisation; neural nets; speech recognition; corrective training; heuristic minimization; hidden control neural network; isolated spoken Korean digit recognition; Artificial neural networks; Backpropagation algorithms; Electronic mail; Error correction; Hidden Markov models; Maximum likelihood estimation; Neural networks; Predictive models; Speech recognition; Testing;
Conference_Titel :
Neural Networks, 1995. Proceedings., IEEE International Conference on
Conference_Location :
Perth, WA
Print_ISBN :
0-7803-2768-3
DOI :
10.1109/ICNN.1995.488189
